Method Details
Details for method 'ShuffleSeg'
Method overview
| name | ShuffleSeg |
| challenge | pixel-level semantic labeling |
| details | ShuffleSeg: An efficient realtime semantic segmentation network with skip connections and ShuffleNet units |
| publication | ShuffleSeg: Real-time Semantic Segmentation Network Mostafa Gamal, Mennatullah Siam, Mo'men Abdel-Razek Under Review by ICIP 2018 |
| project page / code | |
| used Cityscapes data | fine annotations, coarse annotations |
| used external data | ImageNet |
| runtime | n/a |
| subsampling | no |
| submission date | February, 2018 |
| previous submissions |
Average results
| Metric | Value |
|---|---|
| IoU Classes | 58.2887 |
| iIoU Classes | 32.355 |
| IoU Categories | 80.2131 |
| iIoU Categories | 62.178 |
Class results
| Class | IoU | iIoU |
|---|---|---|
| road | 95.579 | - |
| sidewalk | 71.9893 | - |
| building | 85.1398 | - |
| wall | 31.8614 | - |
| fence | 33.7053 | - |
| pole | 39.3765 | - |
| traffic light | 44.0418 | - |
| traffic sign | 51.1458 | - |
| vegetation | 88.7066 | - |
| terrain | 63.8078 | - |
| sky | 92.4633 | - |
| person | 64.4464 | 43.9567 |
| rider | 38.4605 | 19.9293 |
| car | 89.1157 | 79.9923 |
| truck | 36.9667 | 16.1791 |
| bus | 51.0959 | 22.6752 |
| train | 40.895 | 22.2188 |
| motorcycle | 35.8694 | 16.2078 |
| bicycle | 52.8197 | 37.6807 |
Category results
| Category | IoU | iIoU |
|---|---|---|
| flat | 95.421 | - |
| nature | 88.1898 | - |
| object | 46.905 | - |
| sky | 92.4633 | - |
| construction | 84.7124 | - |
| human | 66.4496 | 46.502 |
| vehicle | 87.3501 | 77.8539 |